PDA

View Full Version : 7AFTE Update #3



EvilSpeeder
02-22-2005, 02:26 AM
I did some more tuning tonight. The stock ST clutch really starts to go south at about 13psi. The motor still pulls great, and still no indication of detonation (although I didn't pull the plugs). 13psi is definitely do-able with a new clutch and some better tuning. I think anything past 10 is going to require better ignition control, although I don't really have a solid reason for thinking that. I turned it back down to 10 and worked on part throttle stuff for a while.

Everything looks really good except that the car leans out after a shift. Pulls from low rpm (1000rpm) generate a great AF graph all the way across, but when I do a pull, shift and continue at full throttle, the car leans out for just a second. I'm not sure if the regulator is experiencing some sort of 'lag' or what the deal is. It doesn't seem to correspond to the boost coming back up, but I don't log boost, so thats just my butt dyno talking.

I also got my boost controller hooked up correctly. There was some confusion as to how to hook it up to an external wastegate with two ports, but a call to turboxs got me fixed up. Lag is now reduced and I have much better control over the boost.

EvilSpeeder
02-22-2005, 03:58 AM
Here are some datalogs. The dark blue is throttle position, the light blue is rpm (still need to find another trigger that isn't so dirty) and the green is AFR. Notice the lean spot right after I hit 3rd.

http://evilspeed.com/misc_pics/datalog1.jpg

http://evilspeed.com/misc_pics/datalog2.jpg

nuclearhappines
02-22-2005, 05:16 AM
Are you relying on the FPR or on the 440s for the added fuel ?

I am not a reference on this topic but here's what i think:

1-I think you're rich... 13.5:1 is not lean, especially if you have no boost there.
2-When you let off the gas like that the ECU cuts fuel, i would've imagined it to spike ALOT leaner...
3-Keep in mind a/f ratios are delayed... try to envision if you will the o2 spike to occur sooner (corresponding to where you let off the gas) ... you let off the gas ... it takes time for that air to go through and for the o2 sensor to react to it and for you to see the results...
4-I'd also go out on a limb and say you're missfiring at the top of almost every gear... you see your rpm signal break up and you're a/f goes richer... i think you have a miss

PS: i know you're trying to be safe but i believ 12.5:1 is good ...given that you have a high CR engine you might want to be a bit richer...but i think 11:1 is too rich

try gapping down your plugs? doing some 3rd gear pulls ? and leaning it out a bit at least try to get a flat 11.5:1 ?

of course get second opinions first, but that's my explanation of what's there... don't worry about the spikes... but i'd worry more about how long it takes to jump back to a good a/f... ideally the width of the lean spike should be the same width as the time you are off the gas... but it will show up delayed on the graph

edit: where are you getting your rpm signal from ? IGT and IGF?

-nuke

JGS3SGE
02-22-2005, 05:45 AM
Wow, somebody actually used the header I built long, long ago...

7AFTEs are tough for ignition seeing as the coils inside the cap...not many options.
Like nuke said gappem down... colder plugs too?


Post some graphs with MAP signal if you can...

EvilSpeeder
02-22-2005, 02:48 PM
Are you relying on the FPR or on the 440s for the added fuel ?

I am not a reference on this topic but here's what i think:

1-I think you're rich... 13.5:1 is not lean, especially if you have no boost there.
2-When you let off the gas like that the ECU cuts fuel, i would've imagined it to spike ALOT leaner...
3-Keep in mind a/f ratios are delayed... try to envision if you will the o2 spike to occur sooner (corresponding to where you let off the gas) ... you let off the gas ... it takes time for that air to go through and for the o2 sensor to react to it and for you to see the results...
4-I'd also go out on a limb and say you're missfiring at the top of almost every gear... you see your rpm signal break up and you're a/f goes richer... i think you have a miss

PS: i know you're trying to be safe but i believ 12.5:1 is good ...given that you have a high CR engine you might want to be a bit richer...but i think 11:1 is too rich

try gapping down your plugs? doing some 3rd gear pulls ? and leaning it out a bit at least try to get a flat 11.5:1 ?

of course get second opinions first, but that's my explanation of what's there... don't worry about the spikes... but i'd worry more about how long it takes to jump back to a good a/f... ideally the width of the lean spike should be the same width as the time you are off the gas... but it will show up delayed on the graph

edit: where are you getting your rpm signal from ? IGT and IGF?

-nuke

I still need to get another FPR to replace the stock FPR and lower the base pressure. Right now I've got the AFC leaned way out everywhere, it's not a great situation. I'm probably losing some timing due to that. I haven't messed with timing much, just dialed it back to about 8deg base and left it alone.

Thats a great point about the delay in AF, I hadn't really been taking that into account. You're right about it being rich between shifts. I've got the fpr set up so that the pressure at the '0' point is much higher than stock to help with the leaning out at tip-in. I've likely gone too far with it.

The misfire is a possibility. When I first got the car together I drove it a couple times on standard heat range plugs. It definitely had a noticible miss at the top of each gear. I put colder plugs in it (2 heat ranges) and this seemed to solve the high rpm miss from what I could hear and feel, but it's possible that it's still there to a degree. I really need a cleaner signal to tell for sure. I think right now I'm pulling it from a pin called IGN+. I'll check later tonight or tomorrow.

Lance: I'm a little unclear about the MAP situation. I've got a stock MAP sensor and check valves. It still sees a little bit of boost but the ECU never seems to freak out. Did you ever run it without the check valves or with a toyota 2bar MAP? I'll try gapping down the plugs a little. How far should I go?

Also, even though the coil is in the dist, can't you still interrupt the signal for timing adjustment?

nuclearhappines
02-22-2005, 06:09 PM
when you lean out on the SAFC like that you are advancing your timing .

when i ran with only 1 check valve i did hit fuel cut at around 3psi ... with 2 checkvalves i once spiked past 16psi and didn't see fuel cut ... (pulled like crazy though)

I think what lance was talking about is if you give us a boost vs a/f vs rpm graph it would be informational... that way we can see your boost transition and your a/f ...etc

with a 1 bar the signal off the 1 bar is uselees because it will pretty much go up and cap off at ~4.7 volts as you pass 0psi and stay there till you drop in vacuum again

if you put a 2 bar in there, your corrections on the SAFC will be less but that will limit you to fuel cut at 12psi, unless you use an FCD or a BCC, but even though your corrections seem lower from the SAFC stand point, the change in timing is still the same because there is a correction occuring at the 2bar.

there's also some things that you could try that others haven't yet ... like a 2.5 bar will give you an 18psi (somewhere around there) fuel cut, and remap injectors that are 2.5 times your stock injector size... but then you should retune your safc to do it .... same amount of timing advance though ...

you can't interface an MSD BTM with your distributor?

what's your lowest correction on the SAFC ? like at the higher rpms, in the hi throttle map ? is it still deep in the - side ?

-nuke

EvilSpeeder
02-22-2005, 06:27 PM
>>when you lean out on the SAFC like that you are advancing your timing .

You sure? That seems backwards to me.

>>when i ran with only 1 check valve i did hit fuel cut at around 3psi ... with 2 checkvalves i once spiked past 16psi and didn't see fuel cut ... (pulled like crazy though)

Yeah, I'm running two now. It sees some positive pressure (assuming it's calibrated to the AFC correctly, which is a big assumption) but it doesn't seem to care.

>>I think what lance was talking about is if you give us a boost vs a/f vs rpm graph it would be informational... that way we can see your boost transition and your a/f ...etc

Yeah, I'm working on that. I'll probably just put a GM 2bar map in to monitor/log that.

>>with a 1 bar the signal off the 1 bar is uselees because it will pretty much go up and cap off at ~4.7 volts as you pass 0psi and stay there till you drop in vacuum again

>>if you put a 2 bar in there, your corrections on the SAFC will be less but that will limit you to fuel cut at 12psi, unless you use an FCD or a BCC, but even though your corrections seem lower from the SAFC stand point, the change in timing is still the same because there is a correction occuring at the 2bar.

>>there's also some things that you could try that others haven't yet ... like a 2.5 bar will give you an 18psi (somewhere around there) fuel cut, and remap injectors that are 2.5 times your stock injector size... but then you should retune your safc to do it .... same amount of timing advance though ...

I think I would rather stick with the OE 1bar to run the ECU. Changing sensors complicates the part throttle and driveability issue even more.

>>you can't interface an MSD BTM with your distributor?

I'm pretty sure I can, and I will, but not yet.

>>what's your lowest correction on the SAFC ? like at the higher rpms, in the hi throttle map ? is it still deep in the - side ?

Yeah, my smallest correction is probably somewhere around -32%. There are a lot of -50% values in there.

Snafu
02-22-2005, 07:10 PM
Well, Chris and I had a disgussion a while ago when we were talking about the 3SGTE power primer. When you put in negitive values in, you're actually reducing the "load," not exactly the amount of fuel. When you have a smaller load, ignition timing is increased.

Can we talk about map sensors/check valves a little bit more? I was debating between a 2bar and a check valve, however I'm not sure of the possitive/negitive effects of either.

EvilSpeeder
02-22-2005, 07:48 PM
Well, Chris and I had a disgussion a while ago when we were talking about the 3SGTE power primer. When you put in negitive values in, you're actually reducing the "load," not exactly the amount of fuel. When you have a smaller load, ignition timing is increased.

I see your point but has anybody actually observed this?


Can we talk about map sensors/check valves a little bit more? I was debating between a 2bar and a check valve, however I'm not sure of the possitive/negitive effects of either.

Absolutely! This is a great discussion. We're already skirting the edges of my knowledge on the subject. Hopefully some of the other guys have some more experience with this.

nuclearhappines
02-22-2005, 08:03 PM
I know this isn't a celica map, this is a vr4 map
check out the left side where it says load ...

check out the first map (first link) ... look at the 3500 column, look at the bottom most cell ... this is the cell associated with WOT and 4.7 volts on the map sensor (if we were talking about a celica)... see how it says 9 ?

so total timing advance is base + 9 = 19*

if you're correction at that rpm at WOT is around -50% then you are somewhere in that same column but somewhere in the middle of it ...somewhere like where that 30 is : )

so total timing advance is base + 30 = 40* ...
but you pulled 2 degrees out of base so you are at 38*

what does this mean? this means you are 19* advanced in boost at 3500 rpms
Get my drift ?

do the same comparison at redline and things may not be quite as drastic... basically depends on your corrections and the engine's VE (mostly where it's peak cylinder pressure is ...because timing is highly related to cylinder pressure)

http://www.stealth316.com/images/maxtimingmap-95spydervr4.gif
http://www.stealth316.com/images/redtimingmap-95spydervr4.gif

http://www.stealth316.com/2-ignitionsystem.htm

nuclearhappines
02-22-2005, 08:09 PM
stock map + check valve
2bar

setup 1: stock map + check valve + additonal fuel coming from FPR
Results: no timing advance, limit adjustability, high pressure and high flow fuel pump required (high rail pressure) , no reliable fuel cut, limit boost (no more than 10, 8 is more conservative)

... because the ECU is at the maximum load row on the table because it thinks you are 0psi which is what an n/a car will approach at WOT.

2bar + bigger injectors
Results = some timing advance, still have a fuel cut, low pressure fuel system\
However, if you have your injectors closely matched to your turbo needs (ie at WOT your corrections are small if any +-10%) then your timing advance/retard will be within a few degrees at full boost. It will be advanced at part boost ... but once you reach full fuel flow you're corrections approach 0 and your timing approaches the stock map...and in this case if you have some base retard at the distributor, you will have the same retard compared to the stock map.

i guess the best thing to do evil is to tune it right now as best as you can at the maximum boost you want to run it at and see what your corrections are in your high throttle map and get injectors that are sized accordingly and that will fix most of the timing problem

>>Yeah, my smallest correction is probably somewhere around -32%. There are a lot of -50% values in there.

you need 300cc/min injectors for this setup

-nuke

JGS3SGE
02-23-2005, 02:21 AM
Are you guys installing restictors with those check valves? That makes the check valve work better... I only used 1 CV on my car...

Putting on a 2bar map is no different than the 5SFE 2bar/310cc combo.
You can even use the GM 2bar map sensor (CHEAP) OR the motorola map sensor (even cheaper but a sealed housing will need to made).
You can get an MSD BTW to work with the 7AFE but it requires a special adapter that simulates the coil to fool the smart ignitor (IIRC NO IGF, no spark).

I think the larger map sensor will net much better results than the setup you have now.
Also with the larger map sensor, you can use the TPS signal/map hack which gives boost based correction instead of TPS based corrections (never made sense to me WHY you'd want tps based correction).

Your certainly not being shy with the boost :)

EvilSpeeder
02-23-2005, 06:08 AM
Nuke: Interesting stuff....but theres no way it's advancing that much. It would have died a long time ago if it was. I'm buying your original argument that is is advancing and I acknowledge that you're just using the VR4 map as a reference because thats what you have. I wish we had a similar resource for the celica.

I'm ok with not having a fuel cut. I don't know why you're saying you can't run at higher boost levels though. I understand theres a limit to fuel pressure but thats nothing that bigger injectors can't fix. The setup I'm planning on running soon is a standard aftermarket FPR with the rising rate regulator and the SAFC. This allows me to control fuel in vacuum, at 1atm, and under boost all independently. Small corrections can be made via the SAFC. I definitely think that the 440s are too big for what I'm trying to do right now, but the stock injectors are too small, and 440s is what I have. Sometime soon I might buy some 300s or 310s (anybody have a good source?).

I don't like the idea of using a 2 bar because, instead of lying to the ecu under boost only, it lies to the ecu everywhere. I know, I know...so does the SAFC but I think trying to tune with so many things lying to the ecu is going to get complicated and the 2bar doesn't really get you anywhere.

Lance: Why do you think the MAP route would be better? Tell me more about this TPS signal hack. Are you just replacing the 0-5v TPS signal with a 0-5v pressure based signal?

This is all sounding like it's going to take some dyno time to get right. I'm not too worried about fueling because I have solid feedback on it but the timing issue is worrying me because I have 0 feedback. I don't know what it's doing exactly under any given conditions. I do know that I've seen no signs of detonation which would result from severly advanced timing under boost. I do know where I set my base timing. But that's about all I do know for sure. The settings from the SAFC are very likely causing the ECU to advance timing even further, but we don't know how much.

Thanks for all the input guys, keep it coming.

nuclearhappines
02-23-2005, 07:03 AM
>>I wish we had a similar resource for the celica.

I'm going to start working on cracking all toyota ecu's

>>but I think trying to tune with so many things lying to the ecu is going to get complicated and the 2bar doesn't really get you anywhere.

you're right but you don't know why :p
when you use a higher bar map sensor your resolution drops in half... so from a resolution stand point i'd rather have a 1 bar and take out 20% than have a 2 bar and add 30% to a smaller, dirtier signal.... how much resolution affects your tune i don't know at this point ... would make for interesting research ... probably depends on the ECU's A/D converters (again another thing i can try and figure out when get the ecu)

>>Lance: Why do you think the MAP route would be better? Tell me more about this TPS signal hack. Are you just replacing the 0-5v TPS signal with a 0-5v pressure based signal?

exactly, you are tuning based on load rather than rpm this way, makes your tune more consistent in different gears (which it isn't right now...as you look at gear 3 vs gear 1,2)

>>This is all sounding like it's going to take some dyno time to get right.
Not really, you can street tune it pretty well for most of it since you have a wideband/ logger... i think your street tuning is more meaningful since it has real world load

>>timing issue is worrying me because I have 0 feedback. I don't know what it's doing exactly under any given conditions. I do know that I've seen no signs of detonation which would result from severly advanced timing under boost.

You need to watch EGT's ... if you have your fuel set correctly, then you can start messing with timing (although you don't have control over it right now... )

if your fuel is set, watch your EGT's... if they are too high your timing is either too advanced or too retarded from optimal... egt's will bottom out when your timing is perfect... if egt's are too high and you're on the retarded side you're wasting power...if egt's are too high and you're on the advanced side sooner or later it will detonate...

EGT's will also vary with octane and a/f ratio...but assuming everything else is already fixed (your octane is the same and your a/f is dialed in) then you don't need me to tell you what your timing is doing... it doesn't matter weather it's advanced or retarded...all that matters is weather or not it is safe or not....

you can advance timing under boost... it makes more power... you can add ALOT of timing in boost at places where VE drops... timing advance is not always a bad thing... just make sure you're egt's are safe and will not creep

-nuke

EvilSpeeder
02-23-2005, 02:53 PM
If you crack this ECU you'll be my hero.

Very good point about the resolution of the MAP. I hadn't thought about it from that perspective. I doubt it makes much difference in the end though. Between the resolution of the A/D converters in the ECU and the rate at which the MAP signal changes, it's probably a toss. Remember that you're still dealing with a transient state mechanical device. The engine and the injectors don't respond to the ECU in real time. I wonder how that is taken into account when the ECU is programmed?

Great points about the EGT. I've already got the thermocouple inputs for it, so I'll order the thermocouple today. I don't think the EGT has gone all that high because it hasn't burned the 1200deg paint off the downpipe yet. I fully expected that stuff to last about 50 miles but it's been driven hard for 100s of miles and still hasn't burned off.

A question about EGT probe placement....I understand that it's more responsive and a better representation of whats really goingg on if it's in the manifold but I dont really have a good place to put it that would represent all cylinders. It's much easier to put it in the DP and would more accurately average all cylinders, but would cost me a little bit in response and accuracy. Which location would I be better off with?

ChrisD
02-23-2005, 04:28 PM
On the note of timing advance when you reduce load, well it will definitely occur. By the nature of how engines are tuned, you have more advance if there is less load, mostly. On your question of has it been observed, then yes.

Aaron at ATS who tunes techtom boards is able to watch the changes in ignition advance quite easily by looking at the timing values in each load cell. He posted some guidelines on the mr2 board once, which stated

-Reduce load by 20% ---> timing is advanced by as much as 10 degrees at peak VE

-Reduce load by 10% ---> timing advanced by 2-3 degrees

Now it could be (and probably is) that 10 degrees is the most extra advance you will see. Obviously you are limited within the parameters originally set in the ECU. So you can't see an infinite increase in timing, haha. only the difference in timing between your *actual* load and the load you are telling the ECU to see.

An engine could potentially last quite a while with extra advance, as long as it is running good gas, the a/f is good, and there is no detonation. Being able to tell when detonation sets in, however, is something else entirely!

Also, what wideband is that? The graph display is quite nice.

EvilSpeeder
02-23-2005, 05:13 PM
It's a techedge 2B0.

Linky (http://wbo2.com/2b0/default.htm)

Snafu
02-24-2005, 02:21 AM
Well, seeing as how I don't have an SAFC right now to control fuel (just an Aeromotive FPR/Fuel Pump/310cc injectors), would it be a wise decision to go with the 2 bar map, rather than the check/restrictor valve? It is a rising rate fuel pressure reguator, however I don't believe it would raise fuel enough, or do you think I'll be safe.

Colin

PS. This is just to get it up and running. Once I get it all together, and depending on my talk with some of the pgmfi.org guys, I'll be investing in SMT6 or and SAFC.

Snafu
02-24-2005, 02:24 AM
Oh, and speaking of WB, has anybody looked at this? Is amazing how cheap it is, of course it is a DIY kit.

http://jubjub.mine.nu/wbparts.html

96celica
05-25-2005, 07:03 PM
i know this is old but i'm researching for my 7afte project and just wanted to see how your setup is running

NIK
05-25-2005, 08:12 PM
here is my results with my old 7afte set up
http://www.6gc.net/index.php?action=howto&itu=128
http://www.6gc.net/forums/index.php?showtopic=25034
http://www.6gc.net/forums/index.php?showto...2&hl=7afte+dyno

at 9.5 psi i hit some good numbers 170 whp and 180 trq
it can handle 13-14 psi but i started going lean because of my small injectors

EvilSpeeder
05-25-2005, 08:53 PM
NIK: How much boost did you run on yours? I see you dynoed at 9.5, did you ever run it over that? Thanks

Mine is still running good. I didn't drive it for a couple weeks because it needed a clutch and I had no time to work on it. I've been running about 7psi around town. I need to get the BTM in and some other sensors installed for my data system. I also need to build a cold air box for it. I'm trying to get everything ready for the texas mile in July. I won't be happy unless I can hit the rev limiter in 5th. :D

NIK
05-25-2005, 10:50 PM
the last run on the dyno sheet was one for 14 psi had to cut it at 3500 rpm or so cause i was going lean but got 12 psi at about 186 whp and 215 trq sorry not dyno print out on that one i just sold my set up to a 6gc.net member and am almost done building my 7agte set up looking for 350 whp